Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
12-11-2025 07:15 AM
It sounds like I need to create the "service wrapper" that will do the pre-processing and fetching of env vars etc. I'll deploy that using a model serving endpoint, serverless for speed, then each sub model will be on its own compute cluster that scales independently.
Thanks for the great feedback